AdaAug

AdaAug: Learning class- and instance-adaptive augmentation policies.

Table of Contents

1. Introduction
2. Getting Started
3. Run Search
4. Run Training
5. Citation
6. References & Opensources

Introduction

AdaAug is a framework that finds class- and instance-adaptive data augmentation policies to augment a given dataset.

This repository contains code for the work "AdaAug: Learning class- and instance-adaptive data augmentation policies" (https://openreview.net/forum?id=rWXfFogxRJN) implemented using the PyTorch library.

Getting Started

Code supports Python 3.

Install requirements

pip install -r requirements.txt

Run AdaAug search

Script to search for the augmentation policy for is located in scripts/search.sh. Pass the dataset name as the argument to call the script.

For example, to search for the augmentation policy for reduced_svhn dataset:

bash scripts/search.sh reduced_svhn

The training log and candidate policies of the search will be output to the ./search directory.

Run AdaAug training

To use the searched policy, paste the path of the g_model and h_model as the G and H variables respectively in scripts/train.sh. The path should look like this (./search/...). Then, pass the dataset name as the argument to call the script located in scripts/train.sh. The results will be output to the ./eval directory

bash scripts/train.sh reduced_svhn

Citation

If you use this code in your research, please cite our paper.

@inproceedings{cheung2022adaaug,
  title     =  {AdaAug: Learning class and instance-adaptive data augmentation policies},
  author    =  {Tsz-Him Cheung and Dit-Yan Yeung},
  booktitle =  {International Conference on Learning Representations},
  year      =  {2022},
  url       =  {https://openreview.net/forum?id=rWXfFogxRJN}
}

References & Opensources

Part of our implementation is adopted from the Fast AutoAugment and DADA repositories.

• Fast AutoAugment (https://github.com/kakaobrain/fast-autoaugment)
• DADA (https://github.com/VDIGPKU/DADA)